Research On Hardware Acceleration Technologies Of Detection Algorithms In Multiple-input Multiple-output Systems

As the rapid development of the wireless communication technology, the users, nowadays, require a higher requirement in the scope and speed of the wireless technology. To meet this new requirement, people made the four-generation standards of communicate system, which can support video, Internet input and other operations with higher speed. But the speed still cannot meet the requirement of users in the area of wireless communication system. To support higher speed of data transmitting and users’ motion, the communication system in the next generation must adopt a new transmitting technology with more efficient channel and stronger ability of anti-jamming. In the various feasible solutions at the moment, Multiple-Input Multiple-Output(MIMO) wireless technology is a great breakthrough in modern digital communications domain as the most potential solution.By installing multiple antennas in both of the transmitters and receivers, MIMO system can enhance communication system capacity and spectrum efficiency without increasing bandwidth. MIMO system brings a huge capacity, but at the same time it increases the difficulty of signal detection in the receiver. Thus, in the MIMO wireless communication systems, the receiver signal detection becomes a key issue.In this paper, MIMO detection algriothms are studied in wireless communication filed. Typical algorithms are chosen from four categories of MIMO detection algorithms for the research on the hardware acceleration technologies. They are MMSE detection algorithm from linear dectection algorithms, sorted QR decomposition algorithm from non-linear detection algorithms, SE-based sphere decoding algriothm from maximum likelihood(ML) algorithms, and K-best algriothm from non-ML algriothms. For each algorithm, parallelism is analyzed, basic hardware archithecture is designed, and optimization technologies of hardware acceleration is promoted. Finally, each design is implemented on FPGA or Matlab platform, and performance analysis is done deeply, among which MGS-QRD-based MMSE detector and MMSE-SQRD detector implemented on FPGA platform have fulfilled the real-time detection requirements.